B – 152 Switchaways: a novel embedded validity measure for the Wisconsin Card Sorting Test (WCST)

Abstract Objective We defined a novel WCST validity measure, switchaway errors (SAEs), which occur whenever the examinee correctly matches a card but then switches to a different matching principle for the next item. We assessed whether SAEs can help detect non-credible performance. Method Data were collected from records of 134 adults who underwent private practice neuropsychological evaluations and completed the 128-card WCST and at least one standalone performance validity test. Examinees with ambiguous validity scores or over 64 years old were excluded. The sample (mean age=47.3, education=14.1) included patients diagnosed with mild neurocognitive disorder (n=23), major neurocognitive disorder (n=2), and Attention-Deficit/Hyperactivity Disorder (n=19). Results PVTs revealed invalid performance for 38 patients. Patients with valid performance made an average of 1.61 SAEs, versus 3.16 SAEs for patients with invalid performance (p=.008; Figure 1). A cutoff score of 4 SAEs produced 84.4% specificity and 44.7% sensitivity in identifying which patients had invalid performance (Odds Ratio=4.37), with a statistically significant association (Fisher’s exact test, p<.001). The same analysis for FMS yielded a specificity of 90.6% and a sensitivity of 29.0% at a threshold of 3. Conclusion SAEs yielded similar sensitivity and specificity for identifying invalid performance as FMS, which has previously been linked to validity, suggesting that SAEs could be useful for identifying non-credible performance. Continued data collection will help to increase power and further evaluate these patterns for potential use in clinical practice.